The Swedish BioFINDER 2 Study

GENERAL AIMS:

1. Develop methods for early and accurate diagnosis of different dementia disorders. This is important not only for the clinical diagnostic work-up, but also for selection of patients to clinical trials. Because dementia is very common among the elderly, but often misdiagnosed, we need to develop minimally invasive, reliable and affordable biomarkers for use in a primary care setting. This could include blood-based biomarkers which could be used to identify patients at high risk for a neurodegenerative disease. We also aim to develop new diagnostic algorithms using advanced imaging techniques and cerebrospinal fluid (CSF) biomarkers to diagnose patients prior to overt symptoms (when brain dysfunction is still limited and potentially reversible) in order to identify individuals more likely to respond to new disease-modifying therapies.
2. Develop biomarkers and imaging techniques to monitor early effects of new disease-modifying therapies. Methods are needed that can reliably detect relevant changes in the turnover of Aβ, tau and α-synuclein. In the present study one focus will be to study the annual change in the retention of Tau PET ligands during both the prodromal and dementia stages of AD. Further, we need biomarkers that detect the intensity of ongoing synaptic/neuronal degeneration. Imaging methods revealing the functional and structural integrity of different brain networks might also be relevant.
3. Investigate the heterogeneity of dementia and parkinsonian disorders to assist in the development of a new pathology-based disease classification. Current diagnostic work-up is based on symptomatology. However, the diseases (e.g. Alzheimer's and Parkinson's diseases) are heterogeneous with respect to clinical features and underlying pathologies. Moreover, there is also significant overlap between the diseases. Hence, today's symptom-based clinical diagnostic criteria are likely too crude to provide an etiologically meaningful classification of patients. We will therefore work towards a pathology-based disease classification, using in vivo biomarkers that reflect the underlying brain pathologies, e.g. Aβ or tau. This will be especially useful for the development of new disease-modifying therapies, which are aimed at specific brain pathologies.
4. Define the temporal evolution of pathologies in the predementia phases of Alzheimer's disease. One of the last decade's paradigm shifts in neuroscience has been the realization that AD, and likely also other neurodegenerative diseases, starts with a prolonged predementia phase. AD even starts with an asymptomatic phase, when brain pathology is present in the absence of clinical symptoms. It has become clear that we need to better understand the temporal sequence of pathologic events in these disorders to be able to select the optimal disease stages for interventions in clinical trials with different disease-modifying therapies directed at specific pathologies.
5. Investigate the underlying disease mechanisms of dementia disorders in humans aiming at finding new relevant drug targets. Drug discovery using the currently available cell and animal models, has not translated to human research as indicated by failed phase II and III trials. There are several possible reasons for these failures. First, it is possible that previous trials may have focused on the wrong drug targets, since findings from cell and animal models of dementias may not have accurately captured essential aspects of the disease mechanisms in humans. BioFINDER2 will be a translational study where we will attempt to bridge the knowledge gap between cell/animal studies and studies in humans, by using biomarkers that reflect biological mechanisms that may be studied across model systems. Second, another reason of the failed trials may be that they included patients in too advanced disease stages for the treatments to be effective, or that they partly included patients with unspecific diseases, since they did not use biomarker-based methods for inclusion of participants. BioFINDER2 will inform on the design of future clinical trials by providing detailed data about cognitive and functional changes over time in people with well-defined biomarker-characterized brain pathologies.

STUDY PLAN To reach the objectives above, we include well-characterized and clinically relevant populations of patients with dementia and/or parkinsonian symptoms and healthy individuals. We apply several state of the art methodologies in order to develop new brain imaging techniques, new biomarkers in blood and CSF as well as novel methods of assessing important clinical symptoms.

COGNITIVE TESTING Attention and executive function will be assessed with the Trail Making Test A and B (TMT), Symbol Digit Modalities Test (SDMT), and A Quick Test of cognitive speed (AQT). Visuospatial ability will be measured by two subtests from the Visual Objects and Space Perception (VOSP) battery, incomplete letters and cube analysis. Memory will be assessed with the Free and Cued Selective Reminding Test (FCSRT) in cohorts A and B. It will be complemented with the 10-word delayed recall test from ADAS-cog, including a recognition part. Verbal ability will be evaluated with the animal and letter S fluency tests and the 15-item short version of the Boston Naming Test. Global cognition will be assessed with the Mini-Mental State Examination (MMSE). In cohort A and B, a computerized cognitive battery focusing on memory and attention will also be performed.

ASSESSMENTS OF SYMPTOMS, FUNCTIONAL ABILITIES AND GLOBAL FUNCTION Cognitive symptoms. All subjects will rate his or her memory and attention/executive function in relation to others of the same age according the Brief Anosognosia Scale (BAS). We have also added similar questions to cover the other cognitive domains. These questions have been validated against neuropsychological testing but there are data indicating that self-reported cognitive complaints are only valid in a lesser degree of cognitive impairment. To assess a broader range of cognitive complaints, the Subjective Cognitive Decline questionnaire (SCD-q) will be administered to the research subjects. Subjects from cohorts C, D and E will be assessed with cognitive impairment questionnaire (CIMP-QUEST; filled out by an informant).

Functional ability. This will be evaluated with the informant-based Functional Activities Questionnaire (FAQ) or the Amsterdam IADL scale, both focus on instrumental activities of daily living (IADL) known to be affected early in cognitive decline.

Global function. The global cognitive status will be evaluated using the sum of boxes score from the Clinical Dementia Rating scale (CDR) and the global deterioration scale (GDS).

Behavioral and psychological symptoms in dementia (BPSD). BPSD will be assessed by clinicians using the Neuropsychiatric Inventory - Clinician rating scale (NPI-C) developed by Jeffrey Cummings. Mood and anxiety will be further assessed with the Hospital Anxiety and Depression scale (HADS). Frontal Behavioral Inventory (FBI) will be done in FTD-related conditions.

Quality of Life (QoL). The overall health status will be rated by the subjects using the EQ-5D from Euro-QoL. In demented patients this will also be rated by an informant, spouse or close relative.

Sleep. The presence of REM sleep behavior disorder will be evaluated with a single validated composite question derived from the Mayo Sleep Questionnaire. Sleep quality is assessed with the Sleep Scale from the Medical Outcome Study (MOS).

Cognitive reserve. Premorbid cognition and cognitive reserve is approximated from the Cognitive Reserve Index questionnaire (CRI-q; subitems "Education" and "Working activity", not "Leisure time").

MOTOR ASSESSMENTS Motor aspects are evaluated as part of the sub-study Motor-ACT, which includes detailed motor assessments (including dual tasking): clinical assessments (administered by a physiotherapist), patient-reported outcomes and digital measures (see outcome measures).

Detailed motor assessments are included in the following cohorts: B (older healthy controls), C (SCD/MCI with suspicion of incipient neurocognitive disorder) and E (parkinsonian disorders).

CEREBROSPINAL FLUID (CSF) AND BLOOD SAMPLING AND ANALYZES Lumbar CSF samples will be collected according to a standardized protocol and will follow the principles of the Alzheimer's Association Flow Chart for CSF biomarkers. In short, lumbar puncture will be done between 9-12 am. 20-30 ml of CSF will be collected in Low Binding polypropylene tubes, which are stored on ice for 5-20 min until the CSF samples will be centrifuged (2000g, +4°C, 10 min). Thereafter, the CSF will be aliquoted in ca 1 ml portions into Low Binding polypropylene tubes followed by storage at -80°C until batch analyses.

Plasma collection will be done at the same visit as the lumbar puncture. Blood will be drawn into tubes containing either EDTA (5 x 6 ml tubes) or Lithium heparin (3 x 3 ml tubes) as anticoagulant. After centrifugation (2000g, +4°C, 10 min), plasma samples will be aliquoted into polypropylene tubes and stored at -80°C pending biochemical analyses. Further, EDTA-blood (2 x 6 ml) will also be obtained for genetic DNA analyses.

MAGNETIC RESONANCE IMAGING 3 Tesla MRI (Siemens Prisma) will be done in all study cohorts. A wide variety of magnetic resonance imaging (MRI) techniques will be used to study regional brain volume (three-dimensional magnetization-prepared rapid acquisition with gradient echo (3D MPRAGE)), metabolism (MR spectroscopy (MRS)), structural and functional connectivity of different brain regions (diffusion tensor imaging (DTI) and functional MRI (fMRI)), regional blood flow (arterial spin labeling (ASL)), iron deposition (susceptibility-weighted imaging (SWI)) and the presence of small vessel disease (MPRAGE, SWI and fluid-attenuated inversion recovery (FLAIR)). The protocol will take approximately 60 min to perform. No contrast-enhancing agent will be used.

PET IMAGING

Tau PET. PET imaging of tau aggregates will be done in all the included cohorts at baseline. In the present study, Tau PET imaging will be performed using 18F-RO6958948 developed by Hoffmann-La Roche that will provide the precursor for this PET ligand. This tau PET imaging agent has been shown to accurately detect tau pathology in cases with AD when compared to controls. We will perform a 20-30 min PET scan approximately 60 min post intravenous injection of 18F-RO6958948. The impact of the investigation on clinical diagnostic accuarcy and patient care will be investigated. 18F-RO6958948 has not yet been approved for use in clinical routine practice in Sweden, and can only be used in research studies, such as the present study.

Amyloid PET. PET imaging of Aβ aggregates (including 18F-flutemetamol PET) has been approved for use in clinical routine practice in Sweden. In the present study, 18F-flutemetamol PET will be done in non-demented cases only. In the cases with dementia CSF Aβ will be enough to determine the presence or absence of brain amyloid pathology. However, in the cognitively healthy cases and in the patients with SCD or MCI we are interested in following the spread of amyloid pathology throughout the brain during the preclinical stages of AD and the spatial relationship to tau pathology. Therefore, Amyloid PET will be done according to clinical routine procedures in addition to CSF Aβ measurements in these groups. In the present study Amyloid PET will be performed using 18F-flutemetamol. GE Healthcare will provide the precursor for 18F-flutemetamol. A 20 min scan will be performed between 90-110 min post injection of 18F-flutemetamol.

FDOPA PET FDOPA PET is often used as part of clinical routine examinations of patients with parkinsonism to confirm the diagnosis. Here DaTSCAN will be done according to clinical routine procedures in cases with PD, PDD, DLB, MSA, PSP and CBD to confirm the clinical diagnosis if it has not been done in clinical routine praxis within one year from the baseline visit.